{
\def\sym#1{\ifmmode^{#1}\else\(^{#1}\)\fi}
\begin{tabular}{l*{4}{c}}
\toprule
            &\multicolumn{1}{c}{Compliance}&\multicolumn{3}{c}{Yes-Yes/No-No/Abstain-Abstain}                \\\cmidrule(lr){2-2}\cmidrule(lr){3-5}
            &\multicolumn{1}{c}{\specialcell{All\\(1)}}&\multicolumn{1}{c}{\specialcell{All\\(2)}}&\multicolumn{1}{c}{\specialcell{Below 50th\\(3)}}&\multicolumn{1}{c}{\specialcell{Below 25th\\(4)}}\\
\midrule
& \multicolumn{4}{c}{\textit{Panel A:} Contemporaneous Effect (t)} \vspace{0.2em} \\ \input{pairVoteYesNoAbsA} \vspace{-1.2em} \\ & (.0026) & (.0027) & (.0022) & (.0027) \\ & [.066]* & [.25] & [.019]** & [.25] \\ & \{.037\}** & \{.07\}* & \{.009\}*** & \{.07\}* \\ \\[-5pt] Observations & 35259 & 35259 & 35205 & 35259 \\ \hline \vspace{-8pt} & & & & \\ & \multicolumn{4}{c}{\textit{Panel B:} One Year Later (t+1)} \vspace{0.2em} \\ \input{pairVoteYesNoAbsB} \vspace{-1.2em} \\  & (.0038) & (.0034) & (.0035) & (.0034) \\  & [.37] & [.79] & [.86] & [.79] \\ & \{.26\} & \{.7\} & \{.8\} & \{.7\} \\ \\[-5pt] Observations & 21589 & 21589 & 21589 & 21589 \\ \hline \vspace{-8pt} & & & & \\ & \multicolumn{4}{c}{\textit{Panel C:} Previous Year (Placebo) (t-1)} \vspace{0.2em} \\ \input{pairVoteYesNoAbsC} \vspace{-1.2em} \\  & (.0027) & (.0024) & (.0025) & (.0024) \\  & [.45] & [.74] & [.7] & [.74] \\ & \{.49\} & \{.73\} & \{.72\} & \{.73\} \\ \\[-5pt] Observations & 21638 & 21638 & 21638 & 21638 \\ \hline Session $\times$ Party Pair $\times$ Strata FE & Y & Y & Y & Y \\ Outcome Mean & .57 & .49 & .33 & .49 \\ Outcome S.d. & .13 & .17 & .17 & .17 \\ \toprule \multicolumn{5}{p{14.9cm}}{\footnotesize \textit{Notes:} Each panel shows the estimates from four linear regressions. Compliance is the proportion of times the two MPs in a pair vote the same way in a given session. Yes-Yes/No-No/Abstain-Abstain is the proportion of times the two MPs in a pair both vote yes, or both vote no, or both abstain in a given session. Below 50th/25th votes are votes in which the share of MPs voting the modal vote is less than the median/25th percentile among all votes. Neighbor is a dummy variable equal to one if the MPs in the pair are randomly assigned to sit next to each other during that session. Dyadic-robust standard errors are in parentheses and dyadic-robust p-values are in square brackets. Randomization inference p-values (1000 draws) are in curly brackets. Special sessions and a short session (2017) are excluded. For lead and lag specifications, sessions are also dropped where lead/lag would be a special/short session or a session in a different parliamentary term. Strata FE are dummy variables for whether both MPs in a pair were pre-assigned seats, one MP in a pair was pre-assigned a seat, or neither MP in a pair was pre-assigned a seat. Outcome Mean and Standard Deviation are for the sample included in the Panel A regressions. *** p<0.01, ** p<0.05, * p<0.1.} \\ \end{tabular} }
